Subject: [PATCH] mmio-decode-movz.patch
- Handle the movz instruction in the decoder. This was necessary to get
X working.
Signed-off-by: Chengyuan Li <chengyuan.li@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
Subject: [PATCH] pit-in-hv.patch
- Use ac_timer for PIT interrupt injection (more accurate time keeping)
- Accelerate some critical PIT I/O operations in the hypervisor
(saves 15us per op). Cuts down guest timer interrupt handler execution time
by a third.
- If the domain is scheduled out, make sure that interrupts are not lost
- Add a generic framework for device I/O interception
Signed-off-by: Edwin Zhai <edwin.zhai@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
Fix the synchronization issues between xend and the device model at
startup time.
Initialize the shared page in the hypervisor. Otherwise, the hypervisor
might try to inject spurious interrupts into the guest due to
uninitialized data.
Signed-off-by: Xin B Li <xin.b.li@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
Merge freefall.cl.cam.ac.uk:/auto/groups/xeno-xenod/BK/xen-unstable.bk
into freefall.cl.cam.ac.uk:/auto/groups/xeno/users/iap10/xeno-clone/xen-unstable.bk
Initial checkin of blktap user-land tools. These are fairly experimental,
but a few people have asked to use them. This checkin also includes
Christian's gnbd client library code.
Signed-off-by: andrew.warfield@cl.cam.ac.uk
Reorganise mm.h to split out 32-bit and 64-bit definitions. Fix x86_64
definitions to mask out the bits that we don't care about.
Signed-off-by: keir.fraser@cl.cam.ac.uk
Some functions aren't static and could be (damn C language!).
I tried turning on -Wmissing-prototypes: unfortunately gives warnings
for functions used in asm, which means introducing gratuitous prototypes
for them. Not sure it's worth it.
1) keyhandler.c: keypress_softirq() and do_task_queues() can be static.
2) physdev.c: pcidev_dom0_hidden() can be static.
3) resource.c/resource.h: check_region is deprecated (racy): remove.
4) sched_bvt.c: lots of things can be static.
5) pci/compat.c: not required for Xen.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (authored)
Signed-off-by: ian.pratt@cl.cam.ac.uk
More x86_64 fixes/cleanups. NB. update_va_mapping hypercalls now
accept a virtual address, *not* a "virtual page number".
Signed-off-by: keir.fraser@cl.cam.ac.uk
blktap driver fixes:
Allow the tap to terminate a block device (not need a backend).
Allow it to run in dom0.
Forward control messages to user space as well.
Signed-off-by: akw27@cl.cam.ac.uk
General shadow code cleanup.
Fixed compilation problems when SHADOW_DEBUG is enabled.
Fixed compilation problems when CONFIG_VMX is undefined.
Simplified l1pte_write_fault and l1pte_read_fault.
Name change: spfn => smfn (shadow machine frame numbers).
In general, the terms pfn and gpfn now refer to pages in the
guest's idea of physical frames (which diffs for full shadow
guests). mfn always refers to a machine frame number.
One bug fix for check_pagetable():
If we're using writable page tables
along with shadow mode, don't check the currently writable page table
page -- check its snapshot instead.
Signed-off-by: michael.fetterman@cl.cam.ac.uk
Various hypercall fixes for x86_64.
Main todos: 1. mmu_updates/update_va_mapping hypercalls.
2. map perdomain_pt into Xen address space.
3. exception/interrupt callbacks to guest OS.
4. user-space ring 3 vs. guest-OS ring 3.
Signed-off-by: keir.fraser@cl.cam.ac.uk
Merge labyrinth.cl.cam.ac.uk:/auto/groups/xeno-xenod/BK/xen-unstable.bk
into labyrinth.cl.cam.ac.uk:/auto/groups/xeno/users/iap10/xeno-clone/xen-unstable.bk
Some cleanups:
- Use ENTRIES_PER_L1_PAGETABLE instead of ENTRIES_PER_L2_PAGETABLE
- Use PAGE_SHIFT instead of L1_PAGETABLE_SHIFT
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
This very simple patch is really just cosmetic. It "fixes" xend so that
it doesn't report restart as: "WARNING (XendDomainInfo:1038) Unknown
config field restart" in the xend.log. I'm not sure who should sign off
on this, but the patch is attached.
From: "B.G. Bruce" <bgb@nt-nv.com>
Signed-off-by: ian@xensource.com
Use 1:1 pagetables used for guest physical mode emulation for doing
phys_to_machine_mapping as well.
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
- Detect the CPU speed automatically, rather than having the user
specify a ips in bochsrc. The user can still specify it to override
the auto-detected value.
- Care should be taken to make sure that tsc_per_bx_tick is not too
small or too big. Otherwise, numerical precision issues may result
in the wrong calculation of elapsed time
Signed-off-by: Edwin Zhai <edwin.zhai@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
- Use a more appropriate path for disk images
- Add a 1 GB disk image as an option
From: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
Fix some of the time virtualization issues.
- Compute the elapsed time correctly in the cpu loop
- Try to inject interrupts in the vmexit handler
Signed-off-by: Edwin Zhai <edwin.zhai@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: ian@xensource.com
